Blog to video in n8n: Claude turns every new post in an RSS feed into an AI video Short with music, voiceover and subtitles (free n8n workflow template)

Hey n8n community, if you write a blog or run one for clients, here’s a free n8n workflow template for blog to video: every new post in an RSS feed becomes a 32-second vertical Short, with AI images, AI music, a voiceover and subtitles, all cut to the beat. Claude Opus 5.5 designs each video from the article, and Shotstack generates the pictures and sound, then renders it, ready for TikTok, Instagram Reels or YouTube Shorts.

One-click import from the template gallery: Create 32-second AI video Shorts from RSS blog posts with Claude and Shotstack | n8n workflow template

Or this file: article-shorts-from-rss-v2.json (55.3 KB)

How the workflow works

One Shotstack API key gets you Nano Banana and ElevenLabs in n8n, with no Google or ElevenLabs account. Claude writes the video, and Shotstack runs those models to make it.

First, n8n downloads Shotstack’s guide for AI agents and reads your RSS feed. Then it takes the newest articles, one at a time, and each one becomes a Short:

  1. Claude reads the guide, the article and a short brief in plain English, and writes the whole video as one Shotstack edit. It plans the cuts and writes prompts for 5 Nano Banana 2 images, a voiceover, a music prompt for ElevenLabs, the subtitles and the animated text.
  2. A Code step checks the edit before rendering. Small stuff it fixes itself. Anything bigger goes back to Claude with the exact problem.
  3. The Shotstack node starts one render, and it makes every image and sound inside the node before putting the video together.
  4. If Shotstack rejects the edit or the render fails, Claude reads the error and fixes the edit. It gets two tries.
  5. The MP4 lands in your article-shorts folder, and n8n moves on to the next article.

What it made from the n8n blog

The video above came from n8n’s post on Gemini 4 Argon. Claude opened on “Gemini 4 Argon just dropped”, said n8n is one of the first to integrate it, slammed in the three ways to use it (MCP, the n8n Assistant and agents) and closed on “Read the full article on the n8n blog”. Another article in the feed, 8n’s guide to running AI workflows on NVIDIA DGX Spark with no cloud API, came out as a different video with its own look and colours. Nobody touched either one.

How to set it up

It takes about ten minutes. You need your own n8n, version 2, running on your computer, in Docker or on a server.

  1. Install the Shotstack community node: search for Shotstack in the nodes panel, or go to Settings → Community nodes → Install and enter `@shotstack/n8n-nodes-shotstack`.
  2. Import the workflow from the gallery link above, or open a new workflow, click ⋯ → Import from File and pick the attached file.
  3. Add your Shotstack API key to Make the video and Check the render. You can start a free trial here : Shotack free trial
  4. Add an API key for your AI model to the two Claude steps. Claude Opus 5.5 is the model I’ve had the most success with.
  5. Put your feed’s link and how many articles to make in Your feed. Make an empty article-shorts folder in n8n’s files folder: in Docker that’s /home/node/.n8n-files, and on an npm install it’s .n8n-files in your home folder, so change the path in Save the short.
  6. Hit Execute workflow. Try one article first: each Short takes about 8 minutes.

What it costs and how long it takes

What Cost
Per Short About 8 Shotstack credits, about $1.60. The 5 images are 3.75, the music 3.75, and the voiceover and the render together under 1
Per month Starter ($39 a month) covers about 25 Shorts, Pro ($99) about 60 and Scale ($199) about 120
AI model On top, billed by your AI provider. The first Short costs the most, because Claude reads Shotstack’s whole guide, which then stays cached for an hour
Free trial 10 credits, valid for 30 days: enough for a first Short
Time About 8 minutes a Short: a couple for Claude, the rest generating the pictures and sound, then the render

Ideas

  1. Swap Read the feed for the RSS Feed Trigger, and a Short is made every time a new post goes live.
  2. Run it on a client’s blog and hand back a Short for every article.
  3. Add a YouTube or TikTok step at the end to post each Short.

Good to know

  • It runs on self-hosted n8n, because it saves files on disk. On n8n Cloud, swap the save step for Google Drive or YouTube.
  • I tested it on self-hosted n8n 2.42 with Claude Opus 5.5 and the Shotstack node 0.2.1.
  • The brief is plain English in the step called Claude makes the video. Change the length, the number of images or the style there, or add AI video shots for more motion.
  • Claude only uses facts from the article, and the voiceover follows the article’s language.
  • Use it on your own blog, or one you have permission to repurpose.
  • Read the words before you post a Short anywhere.

Links

If you try it, post your Short in the thread, I’d love to see what your blog turns into. And what should the next one be: translating videos into other languages, or an AI host who presents a new show episode every day?

1 Like

Interesting share. Does self host need certain minimum system requirement?

1 Like

Thanks! Not really. n8n itself is light, so any recent laptop or a small VPS with 2 GB of RAM handles it fine.